The present study was conducted in tertiary care institute of northern India. The medico-legal cases that died of traumatic head injury and brought for autopsy over a period of two years (September 2009 to August 2011) were included in the study. Only those cases were selected who had underwent both X-ray and CT evaluation prior to death. Victims with massive destruction of head and who had surgical intervention were excluded from the study.

A detailed examination and dissection of the head as per standard forensic autopsy procedure was carried out. After dissecting the scalp, temporal muscles and denuding the periosteum the fractures on outer table were noted down. The cranium was opened with an oscillating saw by making a circular cut round the cranium, a little above the eyebrow ridges, keeping close to the reflected flaps of scalp. After removal of the skull cap, the dura was cut with scissors along the line of sawing and reflected. Brain was removed and then the dura mater was stripped from the base of the skull to facilitate its examination for the presence of fractures internally over the base of skull [9].

Apart from the preliminary entries, the radiological reports of skull radiographs and CT scan were collected from the hospital records of the deceased. The data collected was tabulated and comparatively evaluated.

SPSS statistical software version 16.0 was applied to analyse the scientific data. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and accuracy were determined by using 2 by 2 contingency tables for radiological (X-ray and CT) and autopsy, taking autopsy as gold standard and holding CT scan as gold standard while comparing with X-ray.

Forty-two victims of head injury underwent both X-ray and CT evaluation prior to death. Out of 42 X-ray skulls, 20 (47.6%) showed fracture in skull; while during autopsy, fractures were found in 28 (66.7%) subjects indicating that 19.1% fractures were missed on X-ray [Table/Fig-1]. CT showed fractures in 25 cases which signify that only three fracture were missed on CT scan as compared to autopsy. The sensitivity of X-ray for fractures was 71.4%, specificity 100% with accuracy of 80.9%. Kappa was 0.625 which shows good agreement with p value of <0.001 which was statistically highly significant [Table/Fig-2].

On comparison of X-ray and CT, 11.9% fractures were missed on X-ray [Table/Fig-3]. The sensitivity of X-ray for fractures was found to be 80% with accuracy of 88.1%. Kappa was 0.764 which shows good agreement with p value of <0.001 which was statistically highly significant [Table/Fig-4].

It is not possible to detect every single linear fracture of the vault in routine skull radiographic examination. The delineation of a fracture on radiography depends upon the width and direction of the fracture [10]. The temporal bone fractures usually missed on X-ray [11]. Skull films are also suboptimal in revealing of fractures over base of skull [12]. Though plain X-ray can detect skull fractures, they are outdated now. CT scan with bone window is more accurate in detecting depressed skull fractures and more sensitive than skull radiograph [13]. Pfeifer & Pape reviewed several studies for missed injuries in trauma patients and observed that misinterpreted X-rays (15–34.9%) are main radiological factors contributed to missed diagnosis. Clinical inexperience (26.5%), assessment errors (33.8–60.5%), technical errors are additional contributing factors [14]. Cranial sutures are confused with fractures many a times on X-ray. Vascular marking may also be difficult to distinguish from fractures [11].

Although MRI is having more accuracy in diagnosing cerebral pathology, CT is considered the most critical imaging technique for the management of head-injured patients in the acute stage. CT is recommended even for patients with mild head injury having GCS>12 head injury who have the risk factors in form of either loss of consciousness, amnesia, over 60 years of age, seizure, previous neurosurgery, drinker, bleeder, drug abuse. A repeat CT should be done if the initial CT findings were abnormal but the patient’s status has changed within 24 hours after trauma. As the possibility of delayed intracranial haemorrhage cannot be ruled out [8,15]. CT scan also have inherent technical limitations such as beam hardening artefact and partial volume effect which may be responsible for failure to visualize the intracranial lesions. Goyal et al., studied the correlation of CT scan with post-mortem findings of acute head trauma cases. Out of 140 cases taken up for study, 26% of fractures and 8% contusions of brainstem were missed on CT scan when compared with autopsy [16]. Pathak et al., in their study also highlights the fallacies of routine CT scan in detecting lesions close to the bone, e.g. subdural haemorrhage (SDH), subarachnoid haemorrhage (SAH) and basal contusions. Small contusions on the brain stem, cerebral peduncles, corpus callosum or thalamic/hypothalamic areas were also missed in 64% of patients on CT scan. CT scan detected fracture in 13 cases of head injury while it was obvious in 47 cases during autopsy. They observed that the linear fractures at the vault and base of skull are probable to be missed in majority of the cases in routine CT scan and also a normal initial CT scan does not rule out development of delayed intracranial haemorrhage [17]. Yet CT scan is indispensible in evaluation of head trauma patients.

X-ray missed 19.1% of fractures when compared to autopsy, while 11.9% missed when compared to CT scan in our study. It has been observed earlier also by Goel et al., that X-rays detect lesser number of fractures as compared to autopsy findings. Their study concluded that 63.6% of fractured were missed during X-ray when compared with conventional autopsy findings [18]. Thiruppathy et al., also stated that CT has upper hand over plain X-rays in detection of skull fractures [19]. X-ray has lower specificity and accuracy as compared to high-resolution CT scan in delineating skull fractures. The single fracture sensitivity was 71% by conventional CT and 63% by X-ray [20]. Elrahim et al., in their study over 250 patients of head injury concluded that 1.3% linear fracture and 5.1% depressed fracture were missed on X-ray skull. X-ray did not detect single fracture over base of skull which was obvious over CT scan in 12 cases. Plain radiograph was also unable to detect associated intracranial haemorrhages i.e. extradural haemorrhage (EDH), intracerebral haemorrhage, haemorrhagic contusions which were visualized on CT scan [21]. The probability of an intracranial injury increases to fivefold in skull fracture. The meta-analysis by Hofman et al., concluded that plain radiograph is of limited clinical value in analysis of brain injury and its low sensitivity interdict its use [22]. Another study has pointed out that CT scan is more valuable in the evaluation of head injury. Many a times skull fracture is associated with intracerebral haemorrhage which would be improbable to diagnose clinically or by skull radiography alone [23,24].

Yousfani et al., also observed that CT scan had superior performance in precisely assigning the medico legal grade of head injury in contrast to plain X-rays. Out of 100 cases studied by them, X-rays in 21 cases of clinically moderate to severe head injury did not show any injury while CT scans in those cases showed 04 fractures without dislocation, 04 fractures with dislocation, 09 fractures with extradural haemorrhage and 04 fractures of skull with ruptured membranes [25]. A retrospective review of 1845 patients was performed by Masters SJ to evaluate the efficacy of skull films in acute head trauma. The study concluded that skull fracture alone rarely indicates more severe internal head injury. CT scan should be considered as primary diagnostic procedure of choice in cases having features of serious intracranial injury [26]. Increasing use of skull radiography in head injury didn’t yield any useful information and only add up the cost of medical care [27]. In an emergency setting numerous criterion like age and vitals of the patient, pupillary reaction, GCS score are noteworthy predictors for outcome in case of head injury. Still CT scan is long established radiological modality to describe the status of head injury patient. It is efficacious in management of head injury patients in golden hours without wasting valuable time [28].

The detection of skull fracture on conventional radiograph is an indication of serious intracranial injury. However, the diagnostic value of plain radiography is argued. The limited sensitivity of X-ray in detection of skull fracture makes it unreliable. CT is indispensible in the management of acute head injury patients. Skull radiograph is of little benefit when a CT scan is obtained. It has no added advantage over CT scan. Whenever there is facility of CT scan is available, the patient of head injury should not undergo Skull radiograph as it can only delay the diagnosis of an associated intracranial injury and expose the already traumatised patient to harmful radiations.